Full-stack application for document-to-knowledge-graph pipeline: - Backend: FastAPI + LangGraph ReAct agent + DeepSeek + MinerU parsing - Frontend: React 19 + Vite + D3.js + shadcn/ui - Pipeline: MinerU parsing → LangExtract entity extraction → KG building Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
470 B
470 B
Backend — GraphRAG Studio API
路径
F:\GraphRAGAgent\backend\
启动命令
cd F:/GraphRAGAgent/backend
.venv/Scripts/python.exe -m uvicorn main:app --host 0.0.0.0 --port 8000 --reload
接口测试
服务启动后,运行:
.venv/Scripts/python.exe tests/test_api.py
API 文档
- Swagger UI:http://localhost:8000/docs
- ReDoc:http://localhost:8000/redoc
- 健康检查:http://localhost:8000/api/v1/health